Tutorials

Can computer vision and multimodal systems forget?

Machine unlearning is often discussed in the context of privacy – particularly as a response to data removal requests in relation to Europe’s General Data Protection Regulation’s right to be forgotten. However, computer vision and computer science technologists rarely have the opportunity to engage directly with privacy and AI governance policy experts in Machine Unlearning (MU) discus- sions. This tutorial changes that! By bringing together researchers with MU technology expertise and others with privacy and AI governance policy expertise the tutorial aims to improve understanding between both groups. Expect presentations of cutting-edge MU approaches and active Q&A and discussion periods including about policy implications and limitations. Invited speakers include ‘YZ’ Yezhou Yang, associate professor, School of Computing and Augmented Intelligence at Arizona State University and Kairan Zhao, PhD candidate and teaching assistant in Machine Learning at the University of Warwick.

The increasing availability of geospatial data from heterogeneous modalities, including aerial and satellite imagery, ground-level views, and textual descriptions, has made cross-view geo-localization a critical research area with applications in autonomous navigation, urban monitoring, and augmented reality. Despite progress, challenges remain in handling extreme viewpoint variations, scaling across diverse domains, and integrating multimodal information. Recent developments in multimodal learning and Generative AI, such as Large Multimodal Models (LMMs), have introduced new paradigms for geo-localization. LMMs enable more generalized cross-view matching by incorporating language as an additional modality, supporting tasks such as text-based geo-localization, scene description, and multimodal reasoning. These capabilities not only improve performance but also expand the scope of cross-view geo-localization to broader multimodal applications. This tutorial will provide a comprehensive overview of these developments, highlighting the latest methodologies, datasets, and open research directions that are shaping the future of cross-view geo-localization